XB-ART-55302Proc Natl Acad Sci U S A January 1, 2018; 115 (39): E9135-E9144.
Bighead is a Wnt antagonist secreted by the Xenopus Spemann organizer that promotes Lrp6 endocytosis.
The Xenopus laevis embryo has been subjected to almost saturating screens for molecules specifically expressed in dorsal Spemann organizer tissue. In this study, we performed high-throughput RNA sequencing of ectodermal explants, called animal caps, which normally give rise to epidermis. We analyzed dissociated animal cap cells that, through sustained activation of MAPK, differentiate into neural tissue. We also microinjected mRNAs for Cerberus, Chordin, FGF8, BMP4, Wnt8, and Xnr2, which induce neural or other germ layer differentiations. The searchable database provided here represents a valuable resource for the early vertebrate cell differentiation. These analyses resulted in the identification of a gene present in frog and fish, which we call Bighead. Surprisingly, at gastrula, it was expressed in the Spemann organizer and endoderm, rather than in ectoderm as we expected. Despite the plethora of genes already mined from Spemann organizer tissue, Bighead encodes a secreted protein that proved to be a potent inhibitor of Wnt signaling in a number of embryological and cultured cell signaling assays. Overexpression of Bighead resulted in large head structures very similar to those of the well-known Wnt antagonists Dkk1 and Frzb-1. Knockdown of Bighead with specific antisense morpholinos resulted in embryos with reduced head structures, due to increased Wnt signaling. Bighead protein bound specifically to the Wnt coreceptor lipoprotein receptor-related protein 6 (Lrp6), leading to its removal from the cell surface. Bighead joins two other Wnt antagonists, Dkk1 and Angptl4, which function as Lrp6 endocytosis regulators. These results suggest that endocytosis plays a crucial role in Wnt signaling.
PubMed ID: 30209221
Article link: Proc Natl Acad Sci U S A
Genes referenced: angptl4 bmp4 cer1 chrd.1 ctnnb1 ctrl dkk1 egr2 en2 fgf8 frzb h2ac1 h2bc14 lefty loc100494211 loc733561 lrp6 mapk1 mstn.1 nhs nodal2 nodal3.1 nodal3.2 otx2 pkdcc.1 rax sfrp2 sia1 sox2 tf wnt3a wnt8a zic1 zic3 zic4
GO keywords: Wnt signaling pathway
Morpholinos: LOC100494211 MO1 LOC100494211 MO2
Article Images: [+] show captions
|Fig 1 Transcriptome analysis of stage 12 animal cap explants. (A) Illustration of mRNA injection four-cell and animal cap excision at stage (St.) 9. Animal caps were cultured until St. 12 and collected for RNA-seq with or without cell dissociation (Diss). (B) Heat map showing fold changes of the cell dissociation and atypical epidermis signatures. Atypical epidermis corresponds to uninjected animal caps cut at stage 9 and cultured until stage 12; these controls and the dissociated cells correspond to experiment 1 shown in Dataset S1 and are from the same experiment. These gene signatures were also compared with Cerberus, Wnt8, Chordin, FGF8, or BMP4 mRNA-injected animal caps, or with D/V half-embryos derived from the same clutch of embryos. V/D, ventral/dorsal. Rows and columns were left unclustered. Note that many genes induced by dissociation were induced in all conditions, except for BMP4 mRNA-injected animal caps and ventral halves. Genes in the atypical epidermis signature were only induced by BMP4 injection and in ventral halves. A correlation matrix of the animal cap dissociation (C) and atypical epidermis signatures in the RNA-seq libraries (D) are shown. Correlation scores were calculated as Pearson correlation coefficients and color-coded as shown in the scale bar on the right of the panel. These results show that the animal cap signatures obtained via RNA-seq readily distinguish neural- and epidermal-inducing conditions. (E) Dissociation signature examined via PCA to analyze dimensionality in nine experimental conditions (BMP/con AC, Ven/Dor, FGF8/con AC, Xnr2/con AC, Wnt8/con AC, Cer/con AC, Diss/con, Dor/Ven, and Chrd/con AC). AC, animal cap; con, control; Dor, dorsal; Ven, ventral. Each axis represents a principal component (PC1 and PC2), with the first one showing the most variation. PCA clustered epidermal-forming conditions (Ven and BMP), neural-inducing conditions (Diss, Cer, Chrd, and D/V), and mesoderm-forming conditions (FGF8 and Xnr2) without systemic bias. Note that epidermal and dissociation conditions lie on opposing quadrants, indicating the greatest differences. (F) Table of genes induced by Cer mRNA, dissociation, Chrd mRNA, and Wnt mRNA in animal caps listed according to Cerberus fold induction. Because Bighead was induced in all neuralizing conditions and ranked second in the list, it was chosen for further analysis.|
|Fig 2 Bighead is a Spemann organizer-secreted protein. (A) RT-PCR assay showing Bighead (BH) expression across different developmental stages. Histone 4 (H4) was used as a loading control. H2O and –RT served as negative controls. (B) In-situ hybridization of Bighead shows its prominent localization in the Spemann organizer in stage 10 whole embryos and stage 10.5–12 hemisected embryos. Note that there is prominent pan-endodermal localization, particularly in nuclei of the superficial yolk plug region. Embryo pictures were taken at 25× magnification. (C) Wild-type Bighead was secreted into the culture medium by transfected HEK293T cells (lane 5), while a mutant Bighead (ΔBighead) lacking the signal peptide was not (lane 6). Tubulin served as a loading control. IB, immunoblot. (D) Crystal structure of myostatin/GDF8 dimer (38), showing in red the C-terminal part of one of its prodomains that may share structural similarities with Bighead. The amino-terminal part of the myostatin prodomain is shown in gray, and the mature growth factor dimer is shown in yellow. (E) Close-up view of the conserved pro-myostatin structural domain, showing that the deletions (cyan) and insertions (yellow) found in Bighead homologs (Dataset S4) fall within loops without perturbing the β-sheet and α-helical structures.|
|Fig 4 Bighead is required for head development in a BMP-independent way. All embryo pictures were taken at 25× magnification. (A–D) Bighead knockdown inhibits head formation, which is rescued by coinjection of MO-resistant Bighead mRNA. Embryos were injected two times dorsal-marginal at the four-cell stage as indicated and collected at the tailbud stage. The dosages for MO and mRNA were as follows: 32 ng of Bighead (L+S) MOs (directed against the L and S Bighead genes) and 800 pg of Act-HA-Bighead.S mRNA. Numbers of embryos analyzed were as follows: Controls (Ctrl), n = 153, 100% normal; Bighead MOs, n = 122, 93% with a small head phenotype; MO-resistant HA-Bighead mRNA, n = 109, 95% with a dorsalized phenotype; rescue by coinjection of MO and Bighead mRNA, n = 94, 87% rescued. (E and F) In situ hybridization for Otx2 confirming that Bighead knockdown inhibits forebrain development. Embryos were injected two times dorsal-marginal at the four-cell stage with 32 ng of Bighead MO and collected at stage 15. Ctrl, n = 35, 100% normal; Bighead mRNA, n = 48, 83% with enlarged brain phenotype. (G) Expression of the organizer marker chordin at gastrula stage 10.5. (H) Chordin expression was decreased by Bighead MOs. (I) Chordin expression was increased by injection of 800 pg of Bighead mRNA into the animal pole. Note the reduction of Chordin by Bighead MO (n = 34, 100% with phenotype) and expansion of Chordin by Bighead mRNA (n = 47, 93% with increased chordin). (J–O) Experiments with β-catenin–depleted embryos demonstrating that Bighead is not a BMP antagonist but, instead, behaves as a Wnt antagonist. (K) Embryos were injected with 24 ng of β-catenin MO four times into the marginal-vegetal region at the two-cell stage. Then, at the four-cell stage, one dorsal-marginal cell was injected with 1 pg of xWnt8 mRNA (L), 100 pg of Chordin mRNA (M), 50 pg of Dkk1 mRNA (N), or 400 pg of Bighead mRNA (O). Embryos were collected for in situ hybridization with the neural marker Sox2 at stage 18. Note that β-catenin MO completely ventralized embryos and that only the BMP antagonist Chordin could rescue an axis. xWnt8 mRNA was entirely inactive in β-catenin–depleted embryos, as were Dkk1 and Bighead. Numbers of embryos analyzed were as follows: J, n = 47, 100%; K, n = 52, 100% with a ventralized phenotype; L, n = 29, 100% ventralized; M, n = 34, 92% with rescued CNS; N, n = 29, 97% ventralized; O, n = 32, 97% completely ventralized.|